Monday, May 29, 2017

Clock #1 summertime adjustments...

Clock #1 has been in our dining room since I moved it upstairs since March 11.  It ran for several months, and then in mid-April started to become unreliable.  I suspected weather factors were the cause.  After running from mid-April until mid-May, it stopped again, this time apparently for good.  After letting it sit for about 10 days, I went back to investigate the issue.

The overall friction appears to be higher in the spring/summer (though I don't know for certain), although it's unclear exactly if this is localized to a specific mesh in the train..

It seems that the problem was that a few escape teeth were too long by a very small amount (8/1000").  When the friction is lower in the train, the pendulum amplitude is high enough that this doesn't matter.  But when the amplitude drops, this becomes a problem.

To debug the problem, I stuck a post-it note to the back of the frame behind the pendulum.  I marked on this paper the precise pendulum location when each escape tooth released.   The marks were about 13" from the pendulum pivot, and indicate that the typical distance between entry and exit release was 3/16", or about 0.8 degrees peak-to-peak.  So if the pendulum swings less than that, the clock is likely to stop.  I found that one tooth that seemed to cause stoppage consistently corresponded to a mark 1/16" farther out from the rest.  This means that an additional 0.15 degrees was necessary to escape that tooth.  Since the anchor has a length of 1 3/4", this translates to an escape tooth of about (1/16)*(7/4)/13 = 8/1000" longer than the rest.  That's a very small amount, but easily corrected with a file.  Once I corrected that, I checked each other tooth as well, adjusting them so they were all within the 0.4 degree peak-to-peak release margin. 

It seems unlikely that expansion or contraction of the escapement itself (due to weather) is the cause of the clock's malaise, especially because that particular tooth was already marked as being problematic in the past!  Hopefully, my guess about balance amplitude is correct, because it seems to also explain the other issues about the escape wheel sitting at the front/back of the clock too.  This slightly shifts the escape wheel up/down by a very small amount and seems to change the effective length of the escape wheel teeth by a few thousandths of an inch.

Saturday, April 29, 2017

Clearning and reassembly of "La Duchesse"

Now that the pendulum has been replaced for "La Duchesse", it's time to clean and reassemble it.

The movement is a standard two-spring American movement with a count wheel striking mechanism.

Here's another few looks at the movement to make sure I didn't forget anything later upon reassembly!

After taking down the mainsprings (carefully held in place with wire), all of the wheels were removed and cleaned in the ultrasonic cleaner.  The plates were scrubbed with soap and water (avoiding any polishing of anything!) to get the old oil off.  At this point I noticed some interesting things: (1) there are no oil sinks to speak of, (2) there is only minor wear, most of the pivot holes pass the broach test outright, those that didn't were easily opened slightly to round, and (3) the mainsprings were apparently unlubricated.  I also took apart the mainsprings to inspect them -- they were essentially in mint condition.  Dimensions: 3/4" x 0.017" x didn't bother to measure. but they make about 9 full turns when winding

I polished all pivots by hand, using the lathe to hold everything steady.  (No power applied!)

Here are all of the parts cleaned and polished on the bench, ready for reassembly.

Movement reassembly was straightforward except that I had to take several tries to get the warning wheel in the right spot so that striking finished correctly.  Annoyingly, each try displaced all of the pivots, but it wasn't anything major.

The final repair was to fix the door hinge.  The pin was missing, so I made a replacement.  I suspect the original was brass, but I made a new one from steel.  It seems like someone had forced the door downward when it was opened, which cracked off the top hinge pipe on the dial.  I made a replacement one.  The original hinge was clearly soldered in place, but since the dial was painted with gold paint and ceramic, I didn't want to risk heating it.  So instead, I used some sturdy epoxy, which I think was reasonably unobtrusive.  I also had to re-broach the two original hinge pipes because they were warped off in a bad way.  (You can see that in the picture below to some extent.)  Even still, the door doesn't totally snap in place, but at least it closes.

And here is the assembled clock, sitting on my bench.  It's running now and just happily struck 11:00pm.  (However, I must protest Ansonia's choice of an incongruous kitchen-timer-sounding bell for this otherwise somewhat pretentious clock!) 

Monday, April 24, 2017

Replacement pendulum for "La Duchesse"

Since the pendulum on the La Duchesse clock is missing, I am attempting to replace it.  I started by threading a length of 1/8" steel rod with 6-32 threads.  That's at least a standard size for which I could either make a knurled nut or buy one.

I find it easiest to thread by hand on the lathe (with the power off).  That way everything stays neatly aligned.  There are markings on the movement saying "3 3/4", which I take to be roughly the length of the lower rod.  At least, that makes the rod just longer than appears to fit nicely in the case.  It's easier to shorten a rod in any event...

The pendulum rod is segmented.  The upper portion is attached to the suspension spring and the crutch, and is also not missing.  The lower half fits a hook on the upper half.  Judging by how the Ansonia Crystal Regulator was made, the lower rod is flattened at the top and drilled to receive the hook.  So, to that end, I heated and hammer the top of the now-threaded steel rod flat.

I center punched the flat, and tried to drill it with the appropriate size drill.  Since I was using a hand drill (I don't have a drill press, and didn't try to set up the work in the lathe... probably should have), I had trouble with the bit slipping.

So I gave up, and bent a hook in the lower half as well.  This was now easy since the steel was flat and annealed.  However, it was too wind to receive the hook in the top rod.  So I filled the hook in the lower rod to fit the upper hook nicely.

Once filed, I hardened and tempered the rod to blue.

I center drilled and turned a bob from a piece of lead (total weight around 1 oz).

I was initially concerned about turning lead being problematic, but it went nicely enough and the surface finish was good.  Since the bob will be concealed inside the case, I didn't feel a need to attempt to polish it.  Eye protection (as usual) was a must, but especially so since the chips were launched everywhere... messy clean-up.

Sunday, April 23, 2017

Dell Venue 11 pro keyboard fix

My main computer is a Dell Venue 11 pro, which I am generally happy with.  It runs Linux nicely, has an active digitizer stylus, a touch screen, and does just about everything I want.  (The only exception is that it uses micro-HDMI for display output, which is uncommon on projectors... so I can't use it to present at a conference, which is a bummer.)

It's a convertible tablet with a detachable keyboard.  There are various kinds of keyboards: (1) a desktop docking station that I use on my desk at work (I have a USB keyboard, trackball, wired ethernet, and an additional display), (2) a thin folio keyboard that is very lightweight, and (3) a heavier keyboard that also has a spare battery.  The thin keyboard is a little less nice to type on and feels delicate, but I use it when I don't want to carry extra weight.  I usually use the heavier one because typing is nicer.  It basically makes the computer into a laptop.

A few weeks ago, the computer started having issues charging the keyboard's battery.  Worse, it refused to boot at all!  It was quite badly stuck and would not even show the BIOS Dell logo...  Uh oh.

After looking around the forums, I found that the thing to do was to open up the machine, disconnect the battery and the RTC battery (marked), and

then hold the power button for a few seconds.  Upon reassembly, this procedure seemed to restore life to the machine.  Plugging the machine into its dock or the folio keyboard seemed fine, but plugging in the heavier keyboard killed it again!  Figuring that the cause was the keyboard, I ordered a replacement.

The replacement seemed OK except that the spare battery was not detected.  This is apparently a "design feature", and requires you to charge the keyboard separately first.  So I did that, and then trouble struck again...  When the keyboard battery was charged, it bricked the machine again!  Uh oh... evidently the keyboard was not the problem, even though it seemed otherwise.

More searching of forums revealed that the problem is actually that the pins on the bottom of the tablet had retracted into the case and therefore where not making good contact anymore.  The dock uses a separate connector, and is therefore unaffected, and the folio keyboard connector pins are apparently longer.  Since I had to take apart the machine again anyway, I followed the forum post's advice and removed the rest of the case to expose the pins.  This being a modern-ish computer, not only are there screws (with standard heads, thankfully) there are also many delicate-but-stiff spring clips.  I think I managed to avoid breaking them!

It was straightforward to very delicately and gently tap the pins using my staking set from the inside of the case ...

... until they slightly emerged from the bottom of the case as they should.
I reflowed the solder (using lots of flux) around the connector to make sure I hadn't accidentally broken any connections in the process.

This seems to have fixed the problem, as I am posting this using the keyboard and both batteries are indeed showing up as present!

Update (4/24/2017):
When I got to work, the desktop dock didn't work... and later the machine flaked out on the keyboard again.  Argh.  It turns out that the two ribbon cables for the daughterboard at the bottom of the case under the battery (one for the dock, labeled as such on the cable and "DOCK 41 PIN" on the board, leftmost, and the other labeled "LCM" on the cable and "DOCK 45 PIN" on the board, next to it on the right) needed to be reseated.  The cable displacement was not visible (to my unaided eyes) but was visible under a 5x loupe.  In any case, reseating both cables seemed to fix it.  Indeed, the loose ribbon cables might have been the problem all along!

More links about this problem:

Friday, April 21, 2017

More humidity and clock 1

Clock 1 has been mostly reliably running over the past month in our dining room.  It stopped this weekend (April 16) briefly, and then much more extensively today.  Today, it stopped around 8:30am after everyone had left.  I tried to start it again in the afternoon when I got home around 4pm, and spent the next 3 hours fighting with it.  I did find that the escape pinion was slipping -- now fixed -- but it seemed like there was just a general lack of power getting to the escapement.  I couldn't find anything distinctly wrong aside from that, though.

Today was quite warm and humid throughout most of the daylight hours, so this seems to account for the problem.  We had a very short, but very intense thunderstorm around 9-10pm.  After that, the clock started running without any further complications...  The temperature and humidity dropped rather dramatically, so this seems to be anecdotally the cause.

So I went and gathered some historical data to see if that shed any light on the situation.  I got data from and scraped that with a short Python script.

The temperature has been fluctuating quite a bit over the past month.  But it doesn't seem to correlate with either of the clock stopping events.

The air pressure seems mostly irrelevant as well; the clock seems to run fine regardless.

The humidity is really erratic (2 day average in green), like the temperature, though both stoppages did occur with high humidity.

What seems to be causative, though, is the product of the temperature and humidity.  The two largest values (and also the largest deviations from the 2 day averages) were today and over the previous weekend (April 16), both when the clock stopped.

This indicates that the expansion in the wood is really quite small usually, and it takes both factors at work to cause a problem.  More study is surely needed, and I'm sure the weather changing into summer will be interesting...

Sunday, March 26, 2017

Clock 3 escapement adjustments

By using the strategy of topping-within-the-frame, I got the escapement wheel for clock 3 round.  That helped a bit.

By filing the foot spring on the detent thinner, I got it so that 2 lb tangential force on the center wheel got the clock to run.  Still way too much.  So I kept filing, and the spring got very delicate.  I didn't break it, but it was no longer stiff enough to stay vertical.  So, time to make a new detent. 

Figuring that detents were delicate, I made two: a spring one like the one I over-thinned, and a pivoted one. The pivoted detent consists of two parts, a pivot, which attaches to the cock, and the detent itself.  The detent has a counterweight instead of a return spring. Here are the parts, with no springs...

... and here is the pivoted detent assembled. 

It wasn't hard, but required a little thought.  The pivoted detent is much easier to adjust, but even a bit too light.  A few tenths of an ounce on the counterweight seemed to help.  But I broke the pivot hole, so I'll need remake it.

As for the rest of the clock, the idea of a 4-second period for a compound pendulum with a small radius is not going to work.  The longest I can get is about 1.5-2 second period with no escapement in place.  So this means the gearing will be all wrong...  Perhaps this clock too will be a timer.

Sunday, March 19, 2017

Clock 3 update

As suspected in the previous post, the foot spring in the detent is driving the insane need for weight in clock 3.  I thinned it some more, and the required tangential force on the center wheel dropped from 3 lb to 1.5 lb.  So that's heartening.  The detent is now super delicate!

Additionally, since the weight requirement was lessened, the speed of action was reduced.  I therefore caught the escapement fouling... It seems that it's not that the escape wheel teeth are spaced incorrectly.  However, the detent releases the tooth early sporadically.   I think this is made worse as the spring is made weaker, as the detent isn't sitting in its banking.  The early release appears to occur more-or-less consistently at the same escape teeth, so perhaps they're slightly shorter than the others.  In any event, I had filed the detent pallet a bit and now regret doing that!